Debilitating health symptoms brought on by heart disease reduce people’s quality of life and impose serious pain, discomfort, and restrictions on daily activities. It places a heavy burden on economies, healthcare systems, and society at large. Accurate cardiac disease prediction has the ability to significantly contribute to prevention, treatment, and essential assistance for healthcare personnel facing this ailment given its influence on public health. This study uses the most recent developments in machine learning techniques to build an accurate model for heart disease prediction. Heart disease prediction and the Cleveland datasets, which combine approximately 13 important patient history variables, are used to analyze data from people with and without heart disease. XGBoost, naive bayes, logistic regression, decision trees, support vector machines, random forests, and k-nearest neighbors are just a few of the machine learning techniques used in the model development for classification. We can increase the precision and effectiveness of identifying persons at risk of heart disease and enabling prompt therapies by applying these machine learning techniques. According to the findings of this study, XGBoost, decision trees, and random forests have consistently produced high accuracy predictions of heart disease.